CONFORMATION OF METHIONINE
May 20, 1961
2281
volume of Skelly B then was added to the clear filtrate. it slowly became dark brown in color. The reaction mixture was cooled, extracted with 150 ml. of ether, and the While standing inJhe cold overnight, 10 g. of product sepaether phase finally reduced in volume to yield a pre- rated, m.p. 95-96 cipitate which was filtered, recrystallized from absolute Anal. Calcd. for ClzHloNOIBr: C, 51.45; H, 3.67; N, alcoholand dried over calcium chloride in a vacuum desiccator 5.00. Found: C,51.77; H,3.58; N,5.04. to yield 14.7 g. of product, m.p. 75-76'. Ethyl trans-Z-Acetamido-Z-carboethoxy-6-phthalimido-4Anal. Calcd. for C12H10N02Cl: C, 61.15; H, 4.28; N, hexenoate.-A sample of 8 g. of ethyl acetamidomalonate 5.94. Found: C,61.14; H,4.50; N,6.14. was added to a solution of sodium ethoxide prepared from 1 Ethyl cis-2-Acetamido-2-cyano-6-phthalimido-4-hexeno-g. of sodium treated with 100 ml. of magnesium-dried ethate.-A sample of 6.5 g. of ethyl acetamidocyanoacetate was anol, and, after the malonate derivative had dissolved, 10 g. added to a solution of sodium ethoxide prepared from 1.4 g. of N-(trans-4-bromo-2-butenyl)-phthalirnidewas added in of sodium treated with 100 ml. of magnesium-dried ethanol. small increments with frequent shaking. The reaction mixThe The reaction mixture was warmed during the addition to ture then was heated under reflux for about 4 hr. sodium bromide which precipitated during the reaction was effect complete solution of the cyanoacetate, and 10 g. of N-( cis-4-chloro-2-butenyl)-phthalimidethen was added in removed by filtration, and the filtrate was reduced to drysmall increments. After the latter addition was completed, ness in vacuo. The residue was dissolved in a small amount the resulting mixture was heated under reflux for about 2 of ethanol; after which an equal volume of Skelly B was hr. After cooling, the sodium chloride was removed by added, and the resulting solution was placed in the refrigerafiltration and the filtrate was reduced to dryness in vacuo. tor for about one week. There was recovered about 7 g. of product, m.p. 126-127'. The residue was extracted three times with 50 ml. portions of carbon tetrachloride, and the combined organic extract was Anal. Calcd. for C ~ I H ~ ~ NN, ~ O6.72. ~ : Found: N, finally washed with water. The solvent now was removed 6.81. i n vacuo to yield 4.5 g. of residue which was crystallized from trans-2,6-Diamino-4-hexenoicAcid (trans-4,5-Dehydrolyethanol-water to yield 3.5 g. of product, m.p. 97-98'. sine).-A mixture of 7 g. of ethyl trans-2-aceta.mido-2-carAnal. Calcd. for CisHl8N306: N, 11.38. Found: N, boethoxy-6-phthalimido-4-hexenoate aud 50 ml. of concen11.21. trated hydrochloric acid was heated under reflux for about 18 hr. The phthalic acid which formed during the hydrolysis cis-2,6-Diamino-4-hexenoicAcid (cis-4,5-Dehydrolysine). was removed by filtration, and the filtrate was reduced to -A mixture of 3 g. of cis-2-acetamido-2-cyano-6-phthalimido-4-hexenoate and 50 ml. of concentrated hydrochloric dryness in vacuo. The excess hydrochloric acid was removed by repeated addition and evaporation of small quanacid was heated under reflux for about 24 hr. The reaction mixture was cooled, the phthalic acid which precipitated was tities of ethanol i n vacuo. The resulting residue was crysremoved by filtration and the liltrate was reduced to dryness tallized from ethanol-ethyl acetate to yield 700 mg. of prodThe R f values of this compound in After the repeated addition and removal of small uct, m.p. 180-185'. zn vacuo. volumes of ethanol in vacuo, the dried residue was crystal- butano1:acetic acid:water (3:1:1) and 65% pyridine were 0.11 and 0.33, respectively. The ninhydrin spray reagent lized from ethanol-ethyl acetate to yield about 200 mg. of hygroscopic product. The hygroscopic nature of the sample produced a yellow spot which slowly turned purple on standwas such that a melting point could not readily be deter- ing. mined. The material isolated gave a positive test with ninAnal. Calcd. for CaHl*N2O2.2HC1:C, 33.19; H, 6.50; N, hydrin and the Izf values of this product in butano1:acetic 12.90. Found: C, 33.57; H , 6.59; N, 12.66. acid:water ( 3 : l : l ) and 65% pyridine were 0.11 and 0.33, Catalytic Hydrogenation of cis and trans-4,5-Dehpdrolyrespectively. The sample was carefully weighed under anhy- sine.-Both of the cis and trans isomers of 4,5-dehydro-~~drous conditions using a weighing "pig" for an elemental lysine were reduced in the same manner. A solution of 10 analysis. mg. of the appropriate isomer in 10 ml. of water was agitated Anal. Calcd. for C~H1~hT2O2.2HC1: N, 12.90. Found: in the presence of 50 mg. of palladium black under about 45 N, 12.93. lb. of hydrogen pressure for 1hr. The catalyst was removed N-(trans-4-Brorno-Z-butenyl)-phthalimide.-A 40 g. and the appropriate dilutions were made from this 1 mg. per sample of trans-1,4-dibromo-2-butene was heated to its ml. solution for the microbiological assays using Streptomelting point in an oil-bath under anhydrous conditions and, coccus faecalis. The basal medium*was modified by omitting after it had completely melted, 8 g. of potassium phthalimide the DL-lySine, and the assays were supplemented with the was added in small increments with frequent shaking. Upon components indicated and incubated a t 30' for about 20 hr. completion of the addition (about 2 hr.), the reaction mix- The growth response curves obtained with the two hydroture was heated a t about 100-105° for an additional 2 hr. genated samples were quantitatively identical with that The resulting dark brown reaction mixture was extracted found using an authentic sample of DL-lysine. The Rr values of both of the hydrogenated products were identical with ether, and the ether extract was reduced to dryness with a water aspirator. The residue was dissolved in absolute with that of lysine in several paper chromatographic sysethanol, treated with Darco G-60, filtered and an equal tems.
.
[CONTRIBUTION F R O M
THE
CLAYTON FOIJNDATION BIOCHEMICAL INSTITUTE A N D UXIVERSITY O F TEXAS,AUSTIN,TEXAS]
THE
DEPARTMENT
OF
CHEMISTRY, l ' E 1 f C
A Conformation of Methionine Essential for its Biological Utilization BY CHARLES G. SKINNER, JEROME EDELSON AND WILLIAM SHIVE RECEIVED NOVEMBER 11, 1960 The synthesis of the cis and trans forms of 2-amino-4-hexenoic acid (crotylglycine) and of 2-amino-3-methyl-4-pentenoic acid by condensation of the appropriate halide with ethyl acetamidocyanoacetate followed by hydrolysis of the condensation product is accompanied by an allylic rearrangement to give mixtures of crotylglycine and 2-amino-3-methyl-4-pentenoic acid. Biological properties previously ascribed to the trans form of 2-amino-4-hexenoic acid result from a small contamination of rearranged product, and purified samples are ineffective as an amino acid antagonist for Escherichia coli. In contrast, the cis form is a competitive antagonist of methionine for E. coli, so that the conformation of methionine on its site of utilization appears to be one in which the terminal group and the &methylene group are in a cis-like configuration structurally resembling cis- rather than trans-crotylglycine.
Studies of various lysine analogs with restricted rotation have demonstrated that an essential con-
formation of lysine for binding a t its site of utilization is such that the B and e carbons are in a trans-
2282
CH.4RLES
G. SKINNER,
JEROME
EDELSOX AND
Vol. 83
%'ILLIAM SHIVE
C=N
I
e CHs-CH=CH-CHz-C-NHCOCH3
OH -----f
I
cis, x = c i trans, X = Br 1
COOC2H6
i
IV; trans, m.p. 262" dec.
~
CH-CH
\
_.
/CH-cl CHI (racemic mixture)
Fig. 1.-Reaction
CHI'
V,m.p. 117-118"
COOC2Hj
CH*=CH \
(diastereoisomers) CH-CH-COOH
/
I
CH3 NH* VI, m.p. 240-243' dec products of crotyl halides and 3-chloro-1-butene with eth4 1 acetamidocyanoacetate
like configuration. 1--3 Since norleucine is an effective metabolic antagonist of m e t h i ~ n i n e a, ~study of the geometrical isomers of the 4,5-dehydro derivatives of norleucine should give some insight into the conformation of methionine necessary for its biological utilization. In this investigation, both the cis and trans forms of this dehydro derivative, crotylglycine, have been prepared, and thecisisomer but not thetransformhas been found to be an effective competitive antagonist of methionine in Escherichia coli 9723. A preparation of the trans isomer5 has been previously reported to be an amino acid antagonist,6 but this biological activity can now be ascribed to a minor contaminant, 2-amino-3-methyl-4-pentenoic acid, which results from a rearrangement of the halide during the acetamidomalonic ester condensation step. cis-Crotyl alcohol was prepared by catalytic hydrogenation of 2-butyne-1-01 using a previously reported procedure.' The resulting crotyl alcohol was fractionally distilled to give a sample of the cis isomer which on the basis of gas chromatographic analysis contained less than 2% of the trans isomer. The cis-crotyl alcohol was then converted by treatment with phosphorous trichloride to cis-crotyl chloride which was in turn condensed with the sodio derivative of ethyl acetamidocyanoacetate to give ethyl 2-acetamido-2-cyano-cis--2-hexenoate (I). The condensation product subsequently was hydrolyzed with barium hydroxide to yield the desired derivative, 2-amino-cis-4-hexenoic acid (cis-crotylglycine) (11). The reactions are summarized in Fig. 1. h previously reported procedure5was utilized for the preparation of trans-crotylglycine except that fYa7zs-crotyl bromide was used instead of thc cor-
responding chloride for condensation with ethyl acetamidocyanoacetate in alcoholic sodium ethoxide as indicated in Fig. 1. S o evidence of any contamination of the trans-crotyl bromide by the cisisomer was observed using gas chromatographic analysis. Alkaline hydrolysis of the cyanoacetic ester condensation product (111)gave a preparation of trans-crotylglycine (IV) which had antimicrobial activity similar to that previously reported.6 However, subsequent recrystallizations of the preparation greatly diminished its activity such that the concentration necessary for inhibition of growth of E . coli increased from 10-20 y,/ml. to about 160 y/mL These results suggest that the initially observed antimicrobial activity of preparations of trans-crotylglycine was the result of a trace contaminant. During the course of this study, the preparation of 2-amino-3-methyl-4-pentenoic acid (w-dehydroisoleucine) as a possible isoleucine antagonist was undertaken as indicated in Fig. 1. Preliminary studies of the initial product obtained under refluxing conditions suggested that an allylic rearrangement of the unsaturated halide had occurred during the condensation reaction. Secondary allylic halides have been demonstrated to produce rearranged condensation products with nucleophilic reagents such as sodiomalonic ester.P -Alkaline hydrolysis of the oily condensation product gave a mixture of amino acids consisting of approximately 907, 2-amino-4-hexenoic acid (crotyl glycine) (11 or IV) and 10% 2-amino-3-methyl-4-pentenoic acid In an effort to obtain 2-amino-3-niethyl-3-peiitenoic acid (VI) free of crotylglycine, the conden-
(8) R . E . Kepner, S. m'instein and &', C: Young, a h i d , 71, 11.7 (1949). (9) T h e cnmposition n f this misture was iletermined by catalytic hydrogenation t o form the corresponding saturated amino acids, f o l (1) A. I,, Davis, J . 11. Ravel, C. ( > , Skinner and \V, Shive, z l y c h . Biocheui., 76, 139 (1958). lowed by microbiological assays. T h e stimulation of growth of Lnctobaciliiis ovabinosus 17-J under conditions such t h a t the response ( 2 ) -4. L. Davis, C. C:. Skinner :in(l W. Shivc, .tvc/z B i o c h r m Riop h y s . , 87, 88 (1960). t o isoleucine and alloisoleucine are equivalent was used to determine (3) A. I,. Davis, C. G. Skinner and LV. Shixe, 1.A n i . Ch?iir. Soc., 83, the per cent of VI originally present, and the inhibition of growth of 2279 (1961). Escherichia coli 9723 by t h e reduced material (in comparison with norleucine) was used t o determine the per cent. of I1 and/or I V originally (4) W. M . Harding and m'. Shive, J . B i d . Chem., 174, 743 (1948) (,i) H . L. Goering, S. J. Cristol and K Dittmer, J . A m , Ciiein SOL., present. T h e composition was further established by preparing a mixtitre of subsequently isolated 2-amino-3-methyl-4-pentenoicacid 70, 3310 (1948). ( l O 7 J and 2-amino-4-hexenoic acid (90%) and comparing the mix(6) K . Dittmer, A n n X. Y . A c a d . S c i . , 62, 1274 (1950). (7) L. F. Hatch and S . S . Nesbitt, J . A m . Cheiiz. S O L . ,72, 727 ture with t h e isolated reaction products using X-ray pou-der diffrac tion techniques; t h e two powder diffraction patterns were identical. (1950).
May 20, 1961
2283
CONFORMATION OF METHIONINE
sation of 3-chloro-1-butene with ethyl acetamidocyanoacetate was carried out a t lower temperatures, and the condensation product (V) was carefully recrystallized to obtain a pure preparation. Alkaline hydrolysis of the intermediate produced the anticipated 2-amino - 3 -methyl - 4 - pentenoic acids (VI) which upon hydrogenation were found to promote the growth of L. arabinosus under the same conditions and a t the same concentrations as does either isoleucine or alloisoleucine. Since the synthesis of 2-amino-3-methyl-4-pentenoic acid by the above procedure should result in the formation of two diastereoisomers corresponding to the dehydro derivatives of isoleucine and alloisoleucine, the composition of the preparation was determined by hydrogenation of the unsaturated amino acid(s) followed by an assay procedure using Streptococcus faecalis 8043 which requires isoleucine for growth but cannot effectively utilize alloisoleucine. Io The hydrogenated reaction product was only 50% as effective as isoleucine for growth of this organism in contrast to being equally active with either isoleucine or alloisoleucine in promoting growth of L . arabinosus. Thus, it is evident that the preparation of 2-amino-3-methyl-4-pentenoic acid consisted of an equal mixture of the two diastereoisomeric forms, w-dehydroisoleucine and w-dehydroalloisoleucine. dttempts to separate these diastereoisomers by paper chromatographic techniques, by chromatography on a sulfonated-polystyrene resin and by fractional recrystallization of the acetyl derivatives were unsuccessful, even though the latter two techniques were satisfactory for the separation of mixtures of isoleucine and alloisoleucine.11~12 This preparation of 2-amino-3-methyl-4-pentenoic acid inhibits the growth of E . coli in an inorganic salts-glucose medium a t a concentration of approximately 0.6 yiml. The inhibitory activity of the analog is thus sufficiently potent to account for the antimicrobial activity of preparations of trans-crotylglycine which might contain small amounts of this product resulting from an allylic rearrangement during the alkylation step of the preparative procedure. In an effort to determine whether this were so, samples of trans-crotylglycine after successiverecrystallizations were hydrogenated, and the amount of isoleucine present in the resulting mixture was determined using S. faecalis.10 X comparison of the growth inhibiting properties of the corresponding unsaturated fraction using E . coli with that of the ability of the hydrogenated samples to replace isoleucine in growth of S. faecalis is presented in Table I. Since norleucine, which is the reduction product from the main component, crotylglycine, inhibits the response of L. arabinosus to alloisoleucine as indicated in Table 11, it was not (10) A . JIeister, J . B i d . C h e m . , 196, 813 (1952). (11) A Beckman/Spinco Amino Acid Analyzer packed with sulfonated styrene-8% divinylbenzene co-polymer resin prepared for the Model 120 Amino Acid Analyzer was used for t h e column separation study. S. Moore, D. H. Spackman and W. H. Stein, A n a l . Chem., SO, 1185 (1958); D . H. Spackman, W. H. Stein and S Moore, ibid., 30, 1190 (1958). (12) Attempts t o separate t h e acetyl derivatives were patterned after the method of J. P. Greenstein, S. hI. Birnbaum and L. Levintow, “Biochemical Preparations,” Vol. 111, John Wiley a n d Sons, Inc., S e w York, h-. Y . , 1953, p. 84.
possible to determine the amount of the all0 form present in the various recrystallized fractions. However, in view of the difficulty in separating these diastereoisomers of 2-amino-3-methyl-4-pentenoic acid through recrystallization, it is probable that the fractions contain about equimolar quantities of the two forms. TABLEI CONCENTRATIOS OF REARRAXGED PRODUCT IS FRACTIOSALLY RECRYSTALLIZED CROTPLGLYCISE S O .
recrystalliz.
Amount required t o inhibit growth, E . coli, 7;ml.
Isoleucine content in reduced sample,
20
0.8
40
.4 .2 .1
1 3 4 5
%
80 160
TABLEI1 EFFECTOF HIGH LEVELSOF SORLEUCINE o s GROWTH OF Lactobacillus arabinosus TO ISOLEUCINE ASD RESPONSE ALLOISOLEUCISE Concn. of isoleucine or alloisoleucine T/ml.
0 6 8 10 12
Growth response in presence of m-norleucine, mg./ml. -Xone-71-7 7 - 2 AIIO+loAIIOIsoisoIsoism IsoISOleucine leucine leucine Ieucine leucine leucine Galvanometer readings
7
0 38 44 52 55
0 30 40 48 55
0 35 43 51 55
0 5 12 19 25
0 32 40 46 53
0 0 3 6 9
In order to account for the inhibitory effects of the recrystallized samples of trans-crotylglycine (Table I), in which the samples are effective a t about one-half the concentration level as would be anticipated on the basis of their content of both diastereoisomers of 2-amino-3-methyl-4-pentenoic acid, the effect of crotylglycine upon the toxicity of 2-amino-3-methyl-4-pentenoic acid was determined. h supplement of 20 to 60 y/ml. of tram-crotylglycine (which alone a t these concentrations is non-toxic) caused a decrease of 2 to 4-fold in the amount of 2-amino-3-methyl-4-pentenoic acid necessary for inhibition of growth. Thus, i t may be concluded that the toxic effects of trans-crotylglycine which have been observed are due t o a trace contamination of the product formed from an allylic rearrangement during the condensation with the acetamidocyanoacetic ester. For E . ccli, trans-crotylglycine does not appear to have any appreciable inhibitory activity on growth. In addition, the glycyl peptide of trans-crotylglycine did not have any inhibitory effects on the growth of E . coli. X comparable investigation of the crystalline preparation of cis-crotylglycine after catalytic hydrogenation and a microbial assay for the presence of isoleucine using S.faaecalis showed no evidence of any contamination of the anticipated norleucine with isoleucine. However, the crude reaction mixture, after alkaline hydrolysis and hydrogenation of the resulting amino acids, was found to contain about 0.4T0 isoleucine. Thus, an allylic rearrangement also appears to occur during the condensation of cis-crotyl chloride with acetamidocyanoacetate.
CHARLES G. SKINNER, JEROJIE EDELSON .4ND
2284
IO
IC
/N CROTYLGLYCINE
Fig. 2.-scale models of DL-II:ethl~~iii!~e ( i d . i3) a n d cis and trans-DL-crotS-lglS-cine. Tlie angles arid bond lengths used to represent the crotl-lglycines rrere taken from “Tables of Interatomic Distances arid Coniiguraticxi in Molecules and Ions,” Special Publication S o . 11, Tlie Chemical Society, Burlington Housi., London, W . l , 1 9 3 .
As indicated in Table 111, cis-crotylglycine is inhibitory to the growth of E. coli a t a level of about 10 y per ml., and the growth inhibition is reverscd competitively by methionine over more than a tenfold range of increasing concentrations with an irihibition index of approximately 100. 111 view cf the lack of any appreciable toxicity of the trairsisomer, i t appears that the configuration of inethionine on its site of utilization is one iii which tile terminal methyl group and the 6-methylene group are in a cis-like conformation resembling cis-rather than trans-crotylglycine. TABLE 111 REVERSAL OF cis-CROTYLGLYCISE TOXICITY IN Escherichia coli 9723 BY METHIOKIKE~ DL-CL’S-
Crotylglycine, ?/mi.
0 3
I
6
---0 r F
-I J I
Supplement: DL-Xlethionine, r / m l 02 05 1.0 Galrdnometer readings b -r It7
74
il
20
-, zt,
3
39 2
78 20 2 (5 71 1 IL 6i
